WO2016039784A1

WO2016039784A1 - Determining optimum resources for an asymmetric disaster recovery site of a computer cluster

Info

Publication number: WO2016039784A1
Application number: PCT/US2014/062928
Authority: WO
Inventors: Harish B. Kamath; Bhakthavatsala K. NAIDU
Original assignee: Hewlett Packard Enterprise Development Lp
Priority date: 2014-09-10
Filing date: 2014-10-29
Publication date: 2016-03-17

Abstract

Some examples described herein relate to determining optimum resources for an asymmetric disaster recovery site of a computer cluster. In an example, potential resources for a disaster recovery site of a computer cluster may be identified. Configuration information related to the resources deployed at the computer cluster may be determined. An attribute related to a resource type among the resources deployed at the computer cluster may also be determined. The configuration information and the attribute related to the resource type may be used to identify multiple sets of optimum resources, amongst the potential resources, for the disaster recovery site of the computer cluster, wherein the optimum resources in each set are asymmetric to resources deployed at the computer cluster.

Description

DETERM INING OPTIMUM RESOURCES FOR AN ASYMMETRIC DISASTER RECOVERY SITE OF A COMPUTER CLUSTER

Background

[001] Cluster computing evolved as a means of doing parallel computing work in the 1960s. Arguably, one of the primary motivations that led to cluster computing was the desire to link multiple computing resources, which were underutilized, for parallel processing. In fact it may not be incorrect to state that the history of early computer clusters may be tied with the history of early networks. Cluster computing has evolved a long way since then, and many commercial cluster computing products are available in the market today.

Brief Description of the Drawings

[002] For a better understanding of the solution, embodiments will now be described, purely by way of example, with reference to the accompanying drawings, in which:

[003] FIG. 1 illustrates a system for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example; [004] FIGS. 2A to 2C illustrates various examples of an asymmetric set of optimum resources that may be identified for a disaster recovery site of a cluster;

[005] FIG. 3 illustrates a system for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example;

[006] FIG. 4 illustrates a method of for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example; and

[007] FIG. 5 illustrates a system for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example.

Detailed Description of the Invention

[008] A "computer cluster (or cluster)", as used herein, may be defined as a group of computing systems (for example, servers) and other resources (for example, storage, network, etc.) that act like a single system. A computer cluster may be considered as a type of parallel or distributed processing system, which may consist of a collection of interconnected computer systems cooperatively working together as a single integrated resource. In other words, a cluster is a single logical unit consisting of multiple computers that may be linked through a high speed network. A computing system in a cluster may be referred to as a "node". In an example, each node in a cluster may run its own instance of an operating system.

[009] Clusters may be deployed to improve performance and availability since they essentially act as a single, powerful machine. They may provide faster processing, increased storage capacity, and better reliability. [0010] Like all IT systems or resources, a cluster may be susceptible to failure. For instance, in case of a disaster that may be natural (for example, flood, fire, earthquake, hurricane, smog, etc.) or manmade (for example, hacking, human errors, etc.). An enterprise must develop a disaster recovery plan to prepare for such occurrences. A disaster recovery plan may be defined as those processes, policies and/or procedures that enable the recovery or continuation of vital technology infrastructure and systems that may be critical to an organization following a natural or manmade disaster.

[0011] An example disaster recovery plan for a cluster at a production or primary site of an enterprise may involve replicating each of the cluster resources (at the primary site) to a secondary or a disaster recovery site. Such disaster recovery site may be called "symmetric" since it may include all of the resources of the primary site. In the event of a disaster, the resources at the secondary provide infrastructure backup and ensure continuity of IT processes that were impacted at the primary site. However, due to various reasons (for example, cost), it may not be ideal to replicate all resources of a primary site at a secondary site. In such case a disaster recovery site may be operated with an "asymmetric" configuration of resources. The resources at an asymmetric site are not identical to the resources at a primary site. The asymmetricity may be with respect to number of cluster nodes, number of other cluster resources, storage capacity of cluster resources, processing capacity of cluster resources, and the like, at a secondary site. Considering that an asymmetric site may not have the resources of a primary site, it is a challenging task to identify an optimum set of resources for a cluster in an asymmetric disaster recovery site that could provide similar performance, reliability, security, or other desired features, in the event of a failure of the primary site.

[0012] To address this issue, the present disclosure describes various examples for determining optimum resources for an asymmetric disaster recovery site of a computer cluster. In an example, a computing system (for example, a server) may identify potential resources for a disaster recovery site of a computer cluster (at a primary site). The computing system may also determine configuration information related to the resources deployed at the computer cluster and an attribute related to a resource type among the resources deployed at the computer cluster. The computing system may use the configuration information and the attribute related to the resource type to identify multiple sets of optimum resources, amongst the potential resources, for the disaster recovery site of the computer cluster. In an instance, the optimum resources in each set are asymmetric to resources deployed at the computer cluster. The present disclosure may take into account various factors such as, but not limited to, cost, recovery point objective (RPO), recovery time objective (RTO), a single point of failure (SPOF), etc. to identify an optimum set of resources for a disaster recovery site.

[0013] As used herein, the term "optimum" is not intended to mean that the selection is objectively the best or optimal, but rather that the selection was the result of the techniques described herein. Such a selection may be deemed subjectively optimum due to any number of criteria being met.

[0014] FIG. 1 illustrates a system 100 for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example. System 100 may include a computer cluster 102 and a computing system 104. Computing system may be in communication with computer cluster via a computer network 126. Computer network 126 may be a wireless or wired network. Computer network 126 may include, for example, a Local Area Network (LAN), a Wireless Local Area Network (WAN), a Metropolitan Area Network (MAN), a Storage Area Network (SAN), a Campus Area Network (CAN), or the like. Further, computer network 106 may be a public network (for example, the Internet) or a private network (for example, an intranet). [0015] Computer cluster (or cluster) 102, as defined earlier, may be a group of computing systems and other IT infrastructure resources, collectively termed as "resources", that act like a single system. These "resources" may be hardware resources, software resources, or any combinations thereof. For example, hardware resources may include computer systems, computer servers, workstations, storage devices, network or any other computer devices. And, software resources may include operating system software (machine executable instructions), firmware, and/or application software. Such resources may also include virtual machines, virtual servers, storage resources, load balancers, firewalls, etc.

[0016] In an example, resources present in a computer cluster 102 may be classified into various "resource types" according to functions performed by them. For instance, resources that perform processing or related functions may be termed as "compute" resource type, resources that perform network related or ancillary functions may be termed as "network" resource type, resources that perform storage or related functions may be termed as "storage" resource type, resources that provide application services or related processes may be termed as "application" resource type, and so and so forth. In an instance, each of the resource type may be defined by one or more attributes. An "attribute" of a resource type, as used herein, may describe a characteristic of the resource type. For example, CPU usage time and input-output (I/O) usages are some of the non-limiting examples of the "compute" resource type. In another example, type and number of workloads hosted by a "compute" resource type may be considered as attributes. The term "workload", as used herein, may indicate the number of instructions or processes being executed by a processor during a given period or at a particular instant of time. A workload execution may rely on compute and I/O capabilities of a system. A value may be associated with an attribute of a resource type. In an example, computing system may determine a value associated with an attribute of a resource in a computer cluster (for example, 102). [0017] In the example of FIG. 1 , computer cluster 102 may include computing systems 106, 108, and 1 10, storage resources 1 12 and 1 14, and network resource 1 16. Although FIG. 1 shows three computing systems, two storage systems, and one network resource, other examples may include more or less number of these resources.

[0018] A computing system in a cluster may be referred to as a "node". Each of the computer systems (or "nodes") may be any type of computing device capable of reading machine-executable instructions. Examples of a computing system may include, without limitation, a server, a desktop computer, a notebook computer, a tablet computer, and the like.

[0019] Storage resources 1 12 and 1 14 may include a storage device that may be used to store and retrieve data. Some non-limiting examples of storage resources may include a tape drive, a disk drive, a disk array, an optical disc (such as, CD, DVD and Blu-ray disc), a redundant array of independent disks (RAID), a Direct Attached Storage (DAS) device, and a Network Attached Storage (NAS) device.

[0020] Examples of network resource 1 16 may include, without limitation, a network switch, a network router, a virtual switch, a virtual router, a hub, or the like.

[0021] In an example, various resources (for example, computing systems 106, 108, and 1 10, storage resources 1 12 and 1 14, and network resource 1 16) of computer cluster 102 may be communicatively coupled via a computer network. Computer network may be a wireless or wired network. Computer network may include, for example, a Local Area Network (LAN), a Wireless Local Area Network (WAN), a Metropolitan Area Network (MAN), a Storage Area Network (SAN), a Campus Area Network (CAN), or the like. Further, computer network may be a public network (for example, the Internet) or a private network (for example, an intranet).

[0022] In an example, computer cluster may represent a production site or primary site of an organization. Said site may be used by the organization to provide an IT service or application to an internal or external customer. Some non-limiting examples of such services or applications may include an e-commerce service, a database application, and a cloud application.

[0023] Computing system 104 may represent any type of computing device capable of reading machine-executable instructions. Examples of computing system 104 may include, without limitation, a server, a desktop computer, a notebook computer, a tablet computer, a thin client, a mobile device, a personal digital assistant (PDA), a phablet, and the like. In the example of FIG. 1 , computing system 104 may include a resource module 1 18, a configuration module 120, an attribute module 122, and an optimum resource identification module 124. These modules are described later in this document.

[0024] In an example, a user may wish to create a disaster recovery site that ensures continuation of services provided by a computer cluster 102 at a primary site. A disaster recovery site may be present at a remote geographical location relative to its primary site. In an example, however, a disaster recovery site may be co-located with its primary site. In any case, in case a user desires to create a disaster recovery site for a primary site, computing system 104 may identify potential resources that could be deployed at the disaster recovery site. Such potential resources may include hardware resources, software resources, or any combinations thereof. For example, hardware resources may include computer systems, computer servers, workstations, storage devices, network or any other computer devices. And, software resources may include operating system software (machine executable instructions), firmware, and/or application software. These resources may also include virtual resources, such as, but not limited to, virtual machines, virtual servers, virtual network resources, and virtual storage resources. In an example, potential resources may be identified by a user, such as a system administrator.

[0025] Once potential resources for a disaster recovery site have been identified, computing system 104 may determine configuration information related to the resources deployed as part of the computer cluster 102 at the primary site. In the example of FIG. 1 , such configuration information may, for instance, include various configuration details (for example, IP addresses, network ports, etc.) related to computing systems 106, 108, and 1 10, storage resources 1 12 and 1 14, and network resource 1 16 of computer cluster. Computing system 104 may also determine cluster capacity at the primary site. Capacity needs of each workload in terms of CPU and I/O may be defined or computed through heuristics, which may be used in arriving at the cluster capacity.

[0026] Computing system 104 may determine an attribute related to a resource type among the resources deployed as part of the cluster at the primary site. As mentioned above, an "attribute" of a resource type may describe a characteristic of the resource type. For example, CPU usage time and input- output (I/O) usages are some of the non-limiting examples of a "compute" resource type. Other examples may include resource capacity, workload on a resource, and resource dependencies. In an example, computing system may determine a value associated with an attribute of a resource type in the computer cluster.

[0027] Computing system 104 may further determine a parameter or a plurality of parameters for each of the identified multiples sets of optimum resources. Such a parameter(s) or a value associated with such parameter(s) may help a user select an ideal set of optimum resources (for example, amongst multiple such sets) for a disaster recovery site of a cluster. Some non- limiting examples of such parameters may include a cost associated with implementing an identified set of optimum resources, a recovery point objective (RPO) associated with implementing an identified set of optimum resources, a recovery time objective (RTO) associated with implementing an identified set of optimum resources, and a single point of failure (SPOF) associated with implementing an identified set of optimum resources. In an instance, such information may be provided to a user.

[0028] Computing system 104 may also determine a constraint(s) related to a resource type among the resources deployed in a computer cluster at a primary site. An example of such constraints may include workload dependency amongst the resources. In other words, there may be some workloads in a computer cluster at a primary site that may be dependent on each other. Likewise, there may be mutual exclusion among some of the resources. Said differently, there may be some resources in a computer cluster at a primary site that may not work together. Computing system may determine such constraints related to the resources to identify one or more sets of optimum resources for the disaster recovery site of a computer cluster.

[0029] Computing system may use the configuration information related to the cluster at the primary site and at least one attribute related to a resource type to identify one or multiple sets of optimum resources that may be deployed at the disaster recovery site of the computer cluster. Such a set (or sets) of optimum resources may be determined from the potential resources that were identified earlier. As defined earlier, the term "optimum" is not intended to mean that the selection is objectively the best or optimal, but rather that the selection is the result of the techniques described herein. Such a selection may be deemed subjectively optimum due to any number of criteria being met. For example, a set of optimum resources may be considered as optimal from "cost" perspective. Likewise, another set of optimum resources may be considered as optimal if one factors in recovery time objective (RTO) associated with implementing an identified set of resources. Still another set of resources may be considered as optimal if single point of failure (SPOF) is considered as a factor. Similarly, other factors may be considered to determine a set of optimum resources.

[0030] In an instance, the optimum resources in each set are asymmetric to resources deployed at the computer cluster 102. In other words, the resources identified in each set are not identical to the resources of the computer cluster 102 at the primary site. FIGS. 2A to 2C illustrates various examples of an asymmetric set of optimum resources that may be identified for a secondary site (or disaster recovery site) of a cluster at a primary site. These asymmetric set of optimum resources may be determined from a potential set of resources that may be identified earlier.

[0031] Referring to FIG. 2A, let's consider a scenario where a cluster at a primary site may include resources similar to those illustrated in the context of FIG. 1 , i.e. computing systems (for example, 106, 108, and 1 10), storage resources (for example, 1 12 and 1 14), and network resource (for example, 1 16). Computing systems (for example, 106, 108, and 1 10) may host one or more workloads. For instance, computing system 106 may host workload P1 , computing system 108 may host workload P2, and computing system 1 10 may host workloads P3 and P4. Now given a set of potential resources shortlisted for selection of an optimum set (or sets) of resources for a secondary site, a computing system may determine configuration information related to the resources deployed as part of the computer cluster at the primary site and at least one attribute related to said resources. Let's assume that the attribute considered in this example is CPU usage time of computing systems (for example, 106, 108, and 1 10). The CPU usage time may vary across computing systems (for example, 106, 108, and 1 10). Given this scenario, a computing system may use the configuration information related to the cluster at the primary site and CPU usage time of the computer systems, to identify one or multiple sets (example, 202, 204, and 206) of optimum resources that may be deployed at the disaster recovery site of the computer cluster. In an instance, the optimum resources in each set (example, 202, 204, and 206) are asymmetric to resources deployed at the computer cluster. The resources identified in each set (example, 202, 204, and 206) may differ from each other. For instance, each of the identified sets (example, 202, 204, and 206) may include a fewer number of computing systems. Each set of optimum resources (example, 202, 204, and 206) may provide a service that may be similar to the service provided by the cluster resources at the primary site.

[0032] In an example, the computing system may further determine a parameter or a plurality of parameters for each of the identified multiples sets of optimum resources. In the example figures of FIGS. 2A to 2C, such parameters may include a cost associated with implementing an identified set of optimum resources 210, a recovery time objective ( TO) associated with implementing an identified set of optimum resources 212, and a single point of failure (SPOF) associated with implementing an identified set of optimum resources 214.

[0033] In an instance, one or more sets of optimum resources that are identified by the computing system may be provided to a user (for example, by using a display device). Further, a parameter(s) determined by the computing system, for each of the identified sets of optimum resources, may be provided to a user as well. A user may use said information to select a set of optimum resources, amongst the available options, for a disaster recovery site of a computer cluster.

[0034] FIG. 2B illustrates another example of an asymmetric set of optimum resources that may be identified for a secondary site (or disaster recovery site) of a cluster at a primary site. In this example, workload distribution among the computer systems at the primary site may be used as an attribute by the computing system to identify one or multiple sets of optimum resources that may be deployed at the disaster recovery site of the computer cluster. As illustrated in FIG. 2B, workload distribution may vary across computing systems (for example, 106, 108, and 1 10). For instance, computing system 106 may host workload P1 , computing system 108 may host workload P2, and computing system 1 10 may host workloads P3 and P4. Given this scenario, a computing system may use the cluster configuration information and workload distribution amongst the computer systems in the cluster, to identify one or multiple sets of optimum resources (example, 220 and 222) that may be deployed at the disaster recovery site. In an instance, the optimum resources in each set (example, 220 and 222) are asymmetric to resources deployed at the computer cluster. The resources identified in each set (example, 220 and 222) may differ from each other. For instance, each of the identified sets (example, 220 and 222) may include a fewer number of computing systems, and each computer system may include additional workloads (as compared to the primary site) to provide a service that may be similar to the service provided by the cluster resources at the primary site. For example, in one set 220, computer system N1 may include workload P1 and P3, and computer system N2 may include P2 and P4. 35] FIG. 2C illustrates another example of asymmetric set of optimum resources that may be identified for a secondary site of a cluster at a primary site. In this example, storage resource requirement among the computer systems at the primary site may be used as an attribute by the computing system to identify one or multiple sets of optimum resources that may be deployed at the disaster recovery site of the computer cluster. As illustrated in FIG. 2C, storage resource requirement at the primary site may vary across computing systems (for example, 106, 108, and 1 10). For instance, computing system 106 may require storage devices S1 and S2, computing system 108 may require storage device S1 , and computing system 1 10 may require storage resource S2. Given this scenario, a computing system may use the cluster configuration information and storage resource requirement amongst the computer systems in the cluster, to identify one or multiple sets of optimum resources (example, 230 and 232) that may be deployed at the disaster recovery site. In an instance, the optimum resources in each set (example, 230 and 232) are asymmetric to resources deployed at the computer cluster. The resources identified in each set may differ from each other. For instance, each of the identified sets (example, 230 and 232) may include a fewer number of computing systems, and each computer system may be associated with storage devices of different types or capacity (as compared to the primary site) to provide a service that may be similar to the service provided by the cluster resources at the primary site. For example, computer system N1 may be configured with storage devices S1 and S2 in one set of optimum resources 230. In another set 232, computer system N1 may be associated with storage device S1 and computer system N2 may be associated with storage device S2.

[0036] FIG. 3 illustrates a system 300 for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example. In an example, system 300 may be similar to computing system 104 described above. Accordingly, components of system 300 that are similarly named and illustrated in system 104 may be considered similar. In the example of FIG. 3, system 300 may include a resource module 1 18, a configuration module 120, an attribute module 122, and an optimum resource identification module 124. In an example, the aforesaid components of system 300 may be implemented as machine-readable instructions stored on a machine-readable storage medium. The machine- readable storage medium storing such instructions may be integrated with the system 300, or it may be an external medium that may be accessible to the system 300.

[0037] The term "module" may refer to a software component (machine executable instructions), a hardware component or a combination thereof. A module may include, by way of example, components, such as software components, processes, tasks, co-routines, functions, attributes, procedures, drivers, firmware, data, databases, data structures, Application Specific Integrated Circuits (ASIC) and other computing devices. The module may reside on a volatile or non-volatile storage medium and configured to interact with a processor of a computing device.

[0038] Resource module 1 18 may include machine-readable instructions to identify potential resources for a disaster recovery site of a primary cluster site. In an example, resource module may determine said potential resources based on resources present in the cluster at the primary site. In another example, a user may enter information related to potential resources into the system, which may be used by the resource module for making said determination.

[0039] Configuration module 120 may include machine-readable instructions to determine configuration information related to the resources deployed at the primary cluster site. In an example, said configuration information may relate to computing systems (for example, 106, 108, and 1 10), storage resources (for example, 1 12 and 1 14), network resources (for example, 1 16), and other resources deployed at the primary computer cluster.

[0040] Attribute module 122 may include machine-readable instructions to determine an attribute related to a resource type among the resources deployed at the primary cluster site. As mentioned above, an "attribute" of a resource type may describe a characteristic of the resource type. For example, CPU usage time and input-output (I/O) usages are some of the non-limiting examples of a "compute" resource type. In an example, computing system may determine a value associated with an attribute of a resource type in the computer cluster.

[0041] Optimum resource identification module 124 may include machine- readable instructions to consider the configuration information and at least one attribute related to the resource type to identify multiple sets of optimum resources for the disaster recovery site of the primary cluster site. Such a set (or sets) of optimum resources may be determined from the potential resources that were identified earlier. In an instance, the optimum resources in each set are asymmetric to resources deployed at the computer cluster.

[0042] FIG. 4 illustrates a method 400 of for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example. The method 400, which is described below, may be executed on a computer server such as computing systems 104 and 300 of FIGS. 1 and 3 respectively. However, other computing platforms may be used as well. At block 402, potential resources for a disaster recovery site of a computer cluster may be identified. At block 404, configuration information related to the resources deployed at the computer cluster may be determined. At block 406, an attribute related to a resource type among the resources deployed at the computer cluster may be identified. At block 408, the configuration information and the attribute related to the resource type may be used to identify multiple sets of optimum resources, amongst the potential resources, for the disaster recovery site of the computer cluster, wherein the optimum resources in each set are asymmetric to resources deployed at the computer cluster.

[0043] FIG. 5 illustrates a system 500 for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, according to an example. System 500 includes a processor 502 and a machine-readable storage medium 504 communicatively coupled through a system bus. In an example, system 500 may be analogous to system 104 of FIG. 1 or system 300 of FIG. 3. Processor 502 may be any type of Central Processing Unit (CPU), microprocessor, or processing logic that interprets and executes machine-readable instructions stored in machine-readable storage medium 504. Machine-readable storage medium 504 may be a random access memory (RAM) or another type of dynamic storage device that may store information and machine-readable instructions that may be executed by processor 502. For example, machine-readable storage medium 504 may be Synchronous DRAM (SDRAM), Double Data Rate (DDR), Rambus DRAM (RDRAM), Rambus RAM, etc. or storage memory media such as a floppy disk, a hard disk, a CD-ROM, a DVD, a pen drive, and the like. In an example, machine-readable storage medium 504 may be a non-transitory machine-readable medium. Machine-readable storage medium 504 may store monitoring instructions 506, 508, 510, and 512. In an example, instructions 506 may be executed by processor 502 to identify potential resources for a disaster recovery site of a primary cluster site. Instructions 508 may be executed by processor 502 to determine configuration information related to the resources deployed at the primary cluster site. Instructions 510 may be executed by processor 502 to determine an attribute related to a resource type among the resources deployed at the primary cluster site. Instructions 512 may be executed by processor 502 to use the configuration information and the attribute related to the resource type to identify multiple sets of optimum resources, from the potential resources, for the disaster recovery site of the primary cluster site, wherein the optimum resources in each set are asymmetric to resources deployed at the computer cluster. 44] For the purpose of simplicity of explanation, the example method of FIG. 4 is shown as executing serially, however it is to be understood and appreciated that the present and other examples are not limited by the illustrated order. The example systems of FIGS. 1 , 3 and 5, and method of FIG. 4 may be implemented in the form of a computer program product including computer-executable instructions, such as program code, which may be run on any suitable computing device in conjunction with a suitable operating system (for example, Microsoft Windows, Linux, UNIX, and the like). Embodiments within the scope of the present solution may also include program products comprising non-transitory computer-readable media for carrying or having computer-executable instructions or data structures stored thereon. Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer. By way of example, such computer-readable media can comprise RAM, ROM, EPROM, EEPROM, CD-ROM, magnetic disk storage or other storage devices, or any other medium which can be used to carry or store desired program code in the form of computer-executable instructions and which can be accessed by a general purpose or special purpose computer. The computer readable instructions can also be accessed from memory and executed by a processor. 45] It may be noted that the above-described examples of the present solution is for the purpose of illustration only. Although the solution has been described in conjunction with a specific embodiment thereof, numerous modifications may be possible without materially departing from the teachings and advantages of the subject matter described herein. Other substitutions, modifications and changes may be made without departing from the spirit of the present solution.

Claims

Claims:

1 . A method for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, comprising:

identifying potential resources for a disaster recovery site of a computer cluster; determining configuration information related to the resources deployed at the computer cluster;

determining an attribute related to a resource type among the resources deployed at the computer cluster; and

using the configuration information and the attribute related to the resource type to identify multiple sets of optimum resources, amongst the potential resources, for the disaster recovery site of the computer cluster, wherein the optimum resources in each set are asymmetric to resources deployed at the computer cluster.

2. The method of claim 1 , wherein determining the attribute related to the resource type among the resources deployed at the computer cluster comprises:

identifying the attribute related to the resource type among the resources deployed at the computer cluster; and

determining a value of the attribute related to the resource type among the resources deployed at the computer cluster.

3. The method of claim 2, further comprising determining the value of the attribute related to the resource type for each of the resource types in the resources deployed at the computer cluster.

4. The method of claim 1 , further comprising determining a parameter or a plurality of parameters for each of the identified multiples sets of optimum resources.

5. The method of claim 4, further comprising providing the parameter or the plurality of parameters for each of the identified multiples sets of optimum resources to a user.

6. The method of claim 4, wherein the parameter or the plurality of parameters include one of a cost associated with implementing an identified set of optimum resources, a recovery point objective (RPO) associated with implementing an identified set of optimum resources, a recovery time objective (RTO) associated with implementing an identified set of optimum resources, and a single point of failure (SPOF) associated with implementing an identified set of optimum resources.

7. A system for determining optimum resources for an asymmetric disaster recovery site of a computer cluster, comprising:

a resource module to identify potential resources for a disaster recovery site of a primary cluster site;

a configuration module to determine configuration information related to the resources deployed at the primary cluster site;

an attribute module to determine an attribute related to a resource type among the resources deployed at the primary cluster site; and

an optimum resource identification module to consider the configuration information and the attribute related to the resource type to identify multiple sets of optimum resources, from the potential resources, for the disaster recovery site of the primary cluster site, wherein the optimum resources in each set are asymmetric to resources deployed at the computer cluster.

8. The system of claim 7, wherein the potential resources include one of virtual processing resources, virtual network resources, and virtual storage resources.

9. The system of claim 8, wherein the resource type includes one of a computing node, a network resource, a storage resource, an application, and a workload.

10. The system of claim 8, wherein the attribute related to the resource type includes one of CPU usage time, I/O usage, and workload distribution.

1 1 . A non-transitory machine-readable storage medium comprising instructions for an asymmetric disaster recovery site of a computer cluster, the instructions executable by a processor to:

identify potential resources for a disaster recovery site of a primary cluster site; determine configuration information related to the resources deployed at the primary cluster site;

determine an attribute related to a resource type among the resources deployed at the primary cluster site; and

use the configuration information and the attribute related to the resource type to identify multiple sets of optimum resources, from the potential resources, for the disaster recovery site of the primary cluster site, wherein the optimum resources in each set are asymmetric to resources deployed at the computer cluster.

12. The storage medium of claim 1 1 , wherein the instructions to use the configuration information and the attribute related to the resource type further includes instructions to:

provide the multiple sets of optimum resources for the disaster recovery site of the computer cluster to a user for making a selection.

13. The storage medium of claim 12, wherein the resource attributes include resource capacity, workload distribution at various resources, and resource dependencies.

14. The storage medium of claim 12, wherein the instructions to determine the attribute related to the resource type further includes instructions to:

identify a constraint related to the resource type among the resources deployed at the computer cluster; and use the constraint related to the resource type to identify the multiple sets of optimum resources, from the identified potential resources, for the disaster recovery site of the computer cluster.

15. The storage medium of claim 14, wherein the constraint related to the resource type includes one of workload dependency amongst the resources, and mutual exclusion amongst the resources.